The Research Of Network Security Incident Detection Based On Cloud Architecture

The development of computer network has itself characteristics. Nowadays, cloudcomputing network data processing mode has become the theme of network development. Italso brings new opportunities and challenges for network security. At present, the networksecurity problem is still a threat to the healthy development of the network. How to solvethose issues in the era of big data is a very meaningful research subject.Aiming at this problem, this paper presents a network security incident detection systembased on cloud platform. Cloud platform is used here to analysis network large data; Securityincidents are reported to the administrator through the management platform. Such processingmodels will significantly improve the detection efficiency.In order to further improve the efficiency of detection, in the design of algorithm, thispaper researches on the characteristics of entropies, analyzes different applications’ entropiescharacteristic curves, to explore the change rules of them. Propose a new alive entropy-baseddetection approach, which detects the abnormal network traffic based on the values of aliveentropies. The algorithm is described from the alive entropies’ calculation and computingscales in detail.This paper implements the Hadoop cloud computing environment, and according to thecharacteristics of the environment design a complete test programming model, the multiplealgorithm integrated with cloud computing, effectively improves the calculation speed. Acomparative study illustrates that the proposed approach may easily detect the abnormalnetwork traffic with random characteristics.This article makes the cloud platform as a detection system, design and develops ascalable security incident detection system based on cloud platform. In the practical apply;this system also achieves good results both in the algorithms and system testing.